Xnaron Printable Magnetic u-joint option for Rostock Max

[Generic Default]

Nonono, I'm going to put all of my files here regardless of whether I end up making and selling 3d printer kits or not. I don't put any money into designing parts, so I don't expect to make money off of selling STL files or whatever. And to be more clear about the $170 kit I was talking about, the kit would include;

-Effector plate
-Triple extruder (3 nozzle hotend)
- 1 to 3 cold ends (filament pushers)
-Magnetic arms
-All of the necessary electronics

This kit would probably be geared towards people who are making their own delta printers or upgrading Rostocks. It took me so long to get my printer to where it is today as far as capabilities and print quality, and I see market potential for a product that takes the hassle out of hobby 3d printing. Nobody should have to spend as much time fixing and improving their printer as I did!


Right now the problem is that my ball end cups are meant to press fit onto a 7/16th rod, and the press fit will probably not work if your printer setup isn't nearly identical to mine. Plus most people are using 3/8th inch diameter rods.

I attached all of the printable parts in this post, except for the magnetic cup joints. I will work on getting several sizes of those for various press fits tomorrow.

If you print the attached files, make sure to do it relatively slowly at small layer heights. I think I'm using #8 (4mm) hex head cap screws to glue my chrome balls to. I'll try to get more installation instructions and pictures tomorrow.

[Eaglezsoar]

Thanks for what you have uploaded so far.

[neurascenic]

GenDe,

Yes, thank you for your uploads.

I would be interested in your existing cups as I can just run out and get the same brass tubes.

Would also be interested in your sources for the balls and mags.
So that I could follow best your configuration.

What other factors would come into play, as you mentioned others not being set up like yours?


I would be interested in the 3x-truder setup if/when you get there, mag arms, I want ASAP!

Cheers!

[Generic Default]

I should have the ball cups uploaded within 3 hours.

I printed mine with a 0.25mm nozzle out of nylon and they came out perfect with no defects on the ball mating surface. The interior diameter of the ball cup joint where it press fits with the brass tube is 0.454 inches in SolidWorks, but when I printed it the diameter is significantly smaller (so it can press fit onto the brass tube). This discrepancy between designed inside diameter and actual printed diameter is what will cause problems for people since filament type, nozzle size, and machine calibration can all affect the press fit.

My solution is to make several sizes of ball cups, each having slightly different inside diameters. You print each of them until one press fits firmly, then you print 11 more of that type.

[MSURunner]

I should have the ball cups uploaded within 3 hours.

I printed mine with a 0.25mm nozzle out of nylon and they came out perfect with no defects on the ball mating surface. The interior diameter of the ball cup joint where it press fits with the brass tube is 0.454 inches in SolidWorks, but when I printed it the diameter is significantly smaller (so it can press fit onto the brass tube). This discrepancy between designed inside diameter and actual printed diameter is what will cause problems for people since filament type, nozzle size, and machine calibration can all affect the press fit.

My solution is to make several sizes of ball cups, each having slightly different inside diameters. You print each of them until one press fits firmly, then you print 11 more of that type.

And slicer used. But sounds like you have the solution already in place Curious to maybe give those a go!

[Generic Default]

OK, here is a zip file with various ball cup sizes for for 7/16th (0.4375) inch and 3/8th (0.375) inch cups. For use with 3/8th inch chrome steel balls. There are a bunch of STLs in the zip file, with internal diameters graduated in 0.005 inch increments.

I recommend you pick one of the sizes that is slightly larger than your rod to start with. If it's too tight when you press it it on, print the next size up. It it doesn't fit at all, print a few sizes up.

For my 7/16th brass tubes I printed out the 0.454 inch joint.


I got my magnets from the same source xnaron used. This link is also listed on the first page of this thread.
http://shop.revolutionmachines.com/3-8- ... 8-12pc.htm
You can find 3/8th inch chrome steel ball bearings at most local hardware stores like Ace (in the US) near the machine screws in the fastener aisle.


If you aren't in the US, you might be able to get 10mm magnets and balls to work since 10mm is almost the same size as 0.375 inches.

Remember that you want your arms to be as rigid as possible. Rigidity is proportional to the fourth power of the diameter of your tube or rod, so get thicker arms if you can. (0.4375/0.375)^4 = 1.85 times more rigid.
(0.5/0.375)^4 = 3.16 times more rigid.

[neurascenic]

Outstanding! This will be my first upgrade (though I am going to get it running stock first.)

a Vary gracious thanks! to you, xnaron and Werner Berry!

Chris;

[SuperDave42]

Just wanted to throw out a trick i've used in the past to attenuate vibrations in a hollow high rpm rc helicopter tail rotor drive shaft. I would fill the tube about 1/3 to 1/2 with a high viscosity silicon fluid. "Fluidamper" this helped tremendously by being very dynamic. Basically the rotating shaft equally distributed the viscous fluid by means of centrifugal acceleration. The shaft no longer encountered harmonic frequencies and really smoothed the drive train.

Since these magarms are not spinning the effect will be different. Having a viscous liquid weight could make a dramatic difference. If worried about fluid leakage then possibly the arms could be filled with a silicone RTV component. The amount of fill could be determined through testing.

The idea for fine sand filled arms is also a very good one.

Dave

[Eaglezsoar]

Just wanted to throw out a trick i've used in the past to attenuate vibrations in a hollow high rpm rc helicopter tail rotor drive shaft. I would fill the tube about 1/3 to 1/2 with a high viscosity silicon fluid. "Fluidamper" this helped tremendously by being very dynamic. Basically the rotating shaft equally distributed the viscous fluid by means of centrifugal acceleration. The shaft no longer encountered harmonic frequencies and really smoothed the drive train.

Since these magarms are not spinning the effect will be different. Having a viscous liquid weight could make a dramatic difference. If worried about fluid leakage then possibly the arms could be filled with a silicone RTV component. The amount of fill could be determined through testing.

The idea for fine sand filled arms is also a very good one.

Dave

I think it was a user named Flateric who had the sand idea but for him it worked best 1/4 filled with the sand.
One other question. With your cup design is the ball actually slightly touching the magnet?

[Generic Default]

No, the magnet and the ball are separated by several thousandths of an inch. I found that leaving a gap bigger than about 5 thousandths will make the arms more susceptible to detaching during high accelerations, so if you print the cups correctly they should have a gap of about 0.002" to 0.004".


EDIT;

I just checked the arms to make sure they are actually not touching during use. There is a small circle of PTFE lube at the center of each magnet where it is closest to the ball, which indicates that the chrome ball and magnet face are very close but not quite in contact.

[Eaglezsoar]

Thank you for the reply.

[lordbinky]

If you didn't use the epoxy coated magnets, break out the multi-meter and test if you're touching or not.

[SuperDave42]

If you didn't use the epoxy coated magnets, break out the multi-meter and test if you're touching or not.

Great idea !

[bvandiepenbos]

Just wanted to throw out a trick i've used in the past to attenuate vibrations in a hollow high rpm rc helicopter tail rotor drive shaft. I would fill the tube about 1/3 to 1/2 with a high viscosity silicon fluid. "Fluidamper" this helped tremendously by being very dynamic. Basically the rotating shaft equally distributed the viscous fluid by means of centrifugal acceleration. The shaft no longer encountered harmonic frequencies and really smoothed the drive train.

Since these magarms are not spinning the effect will be different. Having a viscous liquid weight could make a dramatic difference. If worried about fluid leakage then possibly the arms could be filled with a silicone RTV component. The amount of fill could be determined through testing.

The idea for fine sand filled arms is also a very good one.

Dave

interesting concept, might be worth a try.
I do think stepper vibrations are transmitted all the way to hot end and could cause roughness in print.
maybe this could dampen it?
thanks for sharing.

[DanielStein]

Hey Guys,

I read this entire thread the other day in preparation to implement magnetic joints however I was thinking of making a few changes. Firstly, I was going to machine the boom ends out of Delrin as I have access to a lathe. Delrin is fairly inexpensive, slippery and is a good all around plastic. It should work for this application quite well. I was thinking about buying case hardened balls and attempting to drill and thread them myself. This would ideally eliminate the joints coming apart which some users are experiencing. Although this may be unnecessary as many users haven't had this problem. I plan to use the lathe to bring the carbon fiber rods into size. Lastly I plan to print the effector, and cheapskate parts.

My questions to fellow Rostock Max owners, and other forum members are:

Would any of you be interested in Delrin boom ends?

Tapped Ball Bearings?

A kit?

If anyone is interested please PM me. I will probably make no more than 5 sets to start unless there is an overwhelming interest. I am looking for people to test these ends/arms and see if they are actually any better than their printed cousins. I haven't yet determined prices/costs. If anyone has any feedback please provide it.

My intent with this project is simply to help others enjoy 3D printing.

Xnaron - This is extraordinary work you have done!

Dan

[Eaglezsoar]

DanielStein, I sent you a PM and I wanted to thank you for making these options available.

[Cleveralias]

Maybe I'm very lucky but I must have 700 hours on my magarms and I've had zero issues.

With that said, I really wanted to use Delrin from the start, and even though my glued ball ends are holding up fine, of course I'd prefer a stud. So you could count me in for a set if you make them.

Thanks!

[Eaglezsoar]

Maybe I'm very lucky but I must have 700 hours on my magarms and I've had zero issues.

With that said, I really wanted to use Delrin from the start, and even though my glued ball ends are holding up fine, of course I'd prefer a stud. So you could count me in for a set if you make them.

Thanks!

I would leave him a PM so he is sure to read it. I also sent him a PM, hopefully a few more will want them so he has incentive to make them.

[neurascenic]

Add me to that list who PMd him. I already started to purchase components for the conversion, but a machined set of caps and real studs... sounds much better.

¢£

[Cleveralias]

Did so. Thanks, Eaglez.

[Eaglezsoar]

Add me to that list who PMd him. I already started to purchase components for the conversion, but a machined set of caps and real studs... sounds much better.

¢£

Great, that's at least three. That is probably enough incentive for him to follow through. Fingers crossed in hope mode.

[SuperDave42]

I'd love to have a high-end quality engineered and manufactured kit for this option.

[SuperDave42]

Has anybody seen these yet ? http://www.tmcmagnetics.com/ball_joint_assembly.html
looks about perfect !

[mhackney]

Yes, and they are >$20 each. I contacted the company to get volume pricing and even in quantities of 1000, it would cost over $100 for a Rostock.

[SuperDave42]

Yes, and they are >$20 each. I contacted the company to get volume pricing and even in quantities of 1000, it would cost over $100 for a Rostock.

ahh bummer..